Sarcopenia is the age-related loss of muscle that is accompanied by reduced strength and physical performance, and the loss of lean mass is a dramatic and universal feature of aging. However, research has shown inconsistent associations between lean mass and outcomes such as physical performance (walking speed, chair stands performance), disability, hospitalization and mortality. At best, lean mass or muscle mass as assessed by existing techniques is weakly associated with functional outcomes; strength and other qualities of muscle are more robust predictors of functional decline. One explanation for such inconsistency is related to problems with the existing tools for assessment of muscle mass: measurement error and/or feasibility issues plague existing techniques including 24-hour urinary excretion of creatinine; dual energy x- ray absorptiometry (DXA); computed tomography (CT); and magnetic resonance (MR) imaging. Given the limitation in existing methods to assess muscle mass, it is clear that a clinically feasible, direct measure of muscle mass will allow for a more complete understanding of the relation between muscle mass and health. A novel measure of total body skeletal muscle mass is the creatine dilution method. Briefly, the total creatine pool size, and thus total body skeletal muscle mass, can be estimated by orally dosing individuals with deuterated creatine (d3-creatine) and then subsequently measuring labeled creatinine (d3-creatinine) in a single urine sample. We posit that total body skeletal muscle mass as measured by this method is related to strength and physical performance, and that individuals with decreased total body skeletal muscle mass will have an increased risk of falls, fractures and functional impairments. We will use traditional methods such as receiver-operator curves, and newer approaches including the net reclassification index and predictiveness curves, to investigate whether total body skeletal muscle mass as determined by the creatine dilution method is superior to appendicular lean mass (from DXA), age, BMI and strength in predicting the risk of falls, fractures, mortality and mobility limitation in older men. We plan to efficiently test these hypotheses via an ancillary study to the ongoing Osteoporotic Fractures in Men (MrOS) study - a large, well-characterized, prospective, multicenter study. We propose to add this measure to the already funded study Visit 4. We will partner with GlaxoSmithKline (GSK), who will donate the labeled creatine dose and complete the assays in the collected urine. Should our hypotheses prove correct, the creatine dilution method for assessing total body skeletal muscle mass would become an critical research tool used to elucidate the pathophysiology of sarcopenia. By providing a highly precise method to monitor response, the creatine dilution method will aid the development of novel agents for reversing age-related muscle loss. This simple test could be used clinically as an alternative to DXA to identify men at risk of physical decline and to refine sarcopenia definitions.